Natural Language Processing Based Instrument for Classification of Free Text Medical Records

Table 1

The data.

Data	Total	Training data	Testing data
Ultrasonography	12.864	7.720	5.144
Liver		1.360	784
Biliary system		1.227	957
Kidney and urinary system		896	536
Gynecologic		1.372	942
Thyroid		1.101	851
Breast		771	531
Vascular Doppler		993	543
X-ray	10.524	5.262	5.262
Chest		849	849
Abdomen		1.057	1.057
Spine		946	946
Limbs		612	612
Esophagus and stomach		1.224	1.224
Large and small bowels		574	574
Endoscopy	1.468	734	734
Total
Document number (ultrasonography, X-ray, endoscopy)	24.856	13.716	11.140